112 citations · 144 across the 40 of their papers we have counts for
43 papers
SEED: Self-Speculative Decoding via Implicit Encoder-Decoder
Hankun Lin, Patrick Pynadath, Ruqi Zhang
Self-speculative decoding accelerates large language model (LLM) inference by drafting tokens from the target model itself, but faces a sharp tradeoff between the quality and cost…
Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement
Yi Ding, Ruqi Zhang
On-policy distillation (OPD) offers dense token-level supervision as an alternative to the sparse outcome-level advantages of reinforcement learning with verifiable rewards (RLVR).…
Gradient-Guided Reward Optimization for Inference-time Alignment
Hankun Lin, Ruqi Zhang
Ensuring the reliability of Large Language Models (LLMs) under distribution drift requires inference-time adaptation. While inference-time alignment methods such as Best-of- and…
Slithering Through Gaps: Capturing Discrete Isolated Modes via Logistic Bridging
Pinaki Mohanty, Ruqi Zhang
High-dimensional and complex discrete distributions often exhibit multimodal behavior due to inherent discontinuities, posing significant challenges for sampling. Gradient-based di…
DIVERSED: Relaxed Speculative Decoding via Dynamic Ensemble Verification
Ziyi Wang, Siva Rajesh Kasa, Ankith M S +6
Speculative decoding is an effective technique for accelerating large language model inference by drafting multiple tokens in parallel. In practice, its speedup is often bottleneck…
Generative Frontiers: Why Evaluation Matters for Diffusion Language Models
Patrick Pynadath, Jiaxin Shi, Ruqi Zhang
Diffusion language models have seen exciting recent progress, offering far more flexibility in generative trajectories than autoregressive models. This flexibility has motivated a…
SARL: Label-Free Reinforcement Learning by Rewarding Reasoning Topology
Yifan Wang, Bolian Li, David Cho +3
Reinforcement learning is critical to improving large reasoning models, but its success relies heavily on verifiable rewards (RLVR), making it hard to use in open-ended domains whe…
Modular Safety Guardrails Are Necessary for Foundation-Model-Enabled Robots in the Real World
Joonkyung Kim, Wenxi Chen, Davood Soleymanzadeh +9
The integration of foundation models (FMs) into robotics has accelerated real-world deployment, while introducing new safety challenges arising from open-ended semantic reasoning a…
Learning Self-Correction in Vision-Language Models via Rollout Augmentation
Yi Ding, Ziliang Qiu, Bolian Li +1
Self-correction is essential for solving complex reasoning problems in vision-language models (VLMs). However, existing reinforcement learning (RL) methods struggle to learn it, as…
One-Step Diffusion Samplers via Self-Distillation and Deterministic Flow
Pascal Jutras-Dube, Jiaru Zhang, Ziran Wang +1
Sampling from unnormalized target distributions is a fundamental yet challenging task in machine learning and statistics. Existing sampling algorithms typically require many iterat…
CANDI: Hybrid Discrete-Continuous Diffusion Models
Patrick Pynadath, Jiaxin Shi, Ruqi Zhang
While continuous diffusion has shown remarkable success in continuous domains such as image generation, its direct application to discrete data has underperformed pure discrete for…
VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models
Qilin Liao, Anamika Lochab, Ruqi Zhang
Vision-Language Models (VLMs) extend large language models with visual reasoning, but their multimodal design also introduces new, underexplored vulnerabilities. Existing multimoda…
DRIFT: Learning from Abundant User Dissatisfaction in Real-World Preference Learning
Yifan Wang, Bolian Li, Junlin Wu +5
Real-world large language model deployments (e.g., conversational AI systems, code generation assistants) naturally generate abundant implicit user dissatisfaction (DSAT) signals,…
Reward-Shifted Speculative Sampling Is An Efficient Test-Time Weak-to-Strong Aligner
Bolian Li, Yanran Wu, Xinyu Luo +1
Aligning large language models (LLMs) with human preferences has become a critical step in their development. Recent research has increasingly focused on test-time alignment, where…
ViLaD: A Large Vision Language Diffusion Framework for End-to-End Autonomous Driving
Can Cui, Yupeng Zhou, Juntong Peng +6
End-to-end autonomous driving systems built on Vision Language Models (VLMs) have shown significant promise, yet their reliance on autoregressive architectures introduces some limi…
Stacey: Promoting Stochastic Steepest Descent via Accelerated -Smooth Nonconvex Optimization
Xinyu Luo, Cedar Site Bai, Bolian Li +3
While popular optimization methods such as SGD, AdamW, and Lion depend on steepest descent updates in either or norms, there remains a critical gap in handli…
VERA: Variational Inference Framework for Jailbreaking Large Language Models
Anamika Lochab, Lu Yan, Patrick Pynadath +2
The rise of API-only access to state-of-the-art LLMs highlights the need for effective black-box jailbreak methods to identify model vulnerabilities in real-world settings. Without…
Accelerating Inference of Discrete Autoregressive Normalizing Flows by Selective Jacobi Decoding
Jiaru Zhang, Juanwu Lu, Xiaoyu Wu +2
Discrete normalizing flows are promising generative models with advantages such as analytical log-likelihood computation and end-to-end training. However, the architectural constra…
Sherlock: Self-Correcting Reasoning in Vision-Language Models
Yi Ding, Ruqi Zhang
Reasoning Vision-Language Models (VLMs) have shown promising performance on complex multimodal tasks. However, they still face significant challenges: they are highly sensitive to…
More is Less: The Pitfalls of Multi-Model Synthetic Preference Data in DPO Safety Alignment
Yifan Wang, Runjin Chen, Bolian Li +7
Aligning large language models (LLMs) with human values is an increasingly critical step in post-training. Direct Preference Optimization (DPO) has emerged as a simple, yet effecti…
Bayesian Computation in Deep Learning
Wenlong Chen, Bolian Li, Ruqi Zhang +1
Bayesian methods have shown success in deep learning applications. For example, in predictive tasks, Bayesian neural networks leverage Bayesian reasoning of model uncertainty to im…
Gradient GA: Gradient Genetic Algorithm for Drug Molecular Design
Chris Zhuang, Debadyuti Mukherjee, Yingzhou Lu +2
Molecular discovery has brought great benefits to the chemical industry. Various molecule design techniques are developed to identify molecules with desirable properties. Tradition…
Making Reliable and Flexible Decisions in Long-tailed Classification
Bolian Li, Ruqi Zhang
Long-tailed classification is challenging due to its heavy imbalance in class probabilities. While existing methods often focus on overall accuracy or accuracy for tail classes, th…
Exploring Non-Convex Discrete Energy Landscapes: An Efficient Langevin-Like Sampler with Replica Exchange
Haoyang Zheng, Hengrong Du, Ruqi Zhang +1
Gradient-based Discrete Samplers (GDSs) are effective for sampling discrete energy landscapes. However, they often stagnate in complex, non-convex settings. To improve exploration,…